gemini-image-generation

To generate images from text prompts using Gemini, you run the provided generate-image.mjs script with your descriptive text and an output path. The skill processes your prompt and saves the resulting image file directly to your workspace.

Yes, you can edit and transform existing image files with a text prompt by running the edit-image.mjs script. You provide the source image and a prompt describing the modifications, and the skill applies the transformation and saves the output.

Yes, you need a valid GEMINI_API_KEY and GEMINI_MODEL_ID configured in your SKILL environment to use Gemini for image generation. You can also optionally set a GEMINI_BASE_URL for custom endpoints.

You can control the aspect ratio and size of generated image outputs by passing the aspectRatio and imageSize parameters in your request. This allows you to configure the generation output quality and format to meet your specific requirements.

The best way to automate text-to-image creation and save files to a workspace path is by executing the generate-image.mjs script. It processes your prompt, saves the image to the specified path, and prints a MEDIA tag for inline rendering.

When generating multiple images from a single prompt, results must be saved to designated workspace paths. The skill handles multi-image generation, but outputs depend on the configured Gemini model and the complexity of your prompt instructions.